ImageGenerationRequest(图片生成请求)
/v1/images/generations 的文档联合模型。网关不会按此类型解析或改写请求，而是把原始 JSON 透明转发给 NewAPI。model 和 prompt 是通用字段；image、seed、组图、提示词优化和水印字段是 Seedream 等供应商扩展；background、moderation、quality、输出压缩等是 OpenAI GPT Image 字段。不同模型支持的字段和值不同，不要把全部可选字段同时发送。请从接口的文生图、图生图、组图或 GPT Image 示例中选择最接近的一套。
{
"model": "doubao-seedream-5-0-260128",
"prompt": "A serene koi pond at sunset, ukiyo-e style.",
"image": [
"https://assets.example.com/reference/koi-pond.jpeg"
],
"size": "1024x1024",
"n": 1,
"seed": 123456789,
"sequential_image_generation": "disabled",
"sequential_image_generation_options": {
"max_images": 4
},
"stream": false,
"guidance_scale": 2.5,
"response_format": "url",
"watermark": false,
"optimize_prompt_options": {
"mode": "standard"
},
"background": "transparent",
"moderation": "low",
"quality": "standard",
"output_compression": 90,
"output_format": "png",
"partial_images": 0,
"style": "vivid",
"user": "user-8f3c2a",
"tools": [
{
"type": "web_search"
}
],
"property1": "string",
"property2": "string"
}